AortaDiff: A Unified Multitask Diffusion Framework For Contrast-Free AAA Imaging

TL;DR

The paper tackles contrast-free abdominal aortic aneurysm assessment by introducing AortaDiff, a unified multitask diffusion framework that jointly translates NCCT to synthetic CECT and segments the aortic lumen and thrombus. It leverages a shared encoder-decoder with dual heads, operates initialization-free, and utilizes semi-supervised learning to handle missing segmentation labels, evaluated on the OxAAA dataset. Results show state-of-the-art performance across synthesis (PSNR), segmentation (lumen/thrombus Dice), and clinically relevant measurements (lumen diameter, thrombus area), outperforming single-task and multi-stage baselines. This work enables safer, more data-efficient AAA evaluation by providing a digital contrast solution with improved anatomical fidelity and quantitative analysis.

Abstract

While contrast-enhanced CT (CECT) is standard for assessing abdominal aortic aneurysms (AAA), the required iodinated contrast agents pose significant risks, including nephrotoxicity, patient allergies, and environmental harm. To reduce contrast agent use, recent deep learning methods have focused on generating synthetic CECT from non-contrast CT (NCCT) scans. However, most adopt a multi-stage pipeline that first generates images and then performs segmentation, which leads to error accumulation and fails to leverage shared semantic and anatomical structures. To address this, we propose a unified deep learning framework that generates synthetic CECT images from NCCT scans while simultaneously segmenting the aortic lumen and thrombus. Our approach integrates conditional diffusion models (CDM) with multi-task learning, enabling end-to-end joint optimization of image synthesis and anatomical segmentation. Unlike previous multitask diffusion models, our approach requires no initial predictions (e.g., a coarse segmentation mask), shares both encoder and decoder parameters across tasks, and employs a semi-supervised training strategy to learn from scans with missing segmentation labels, a common constraint in real-world clinical data. We evaluated our method on a cohort of 264 patients, where it consistently outperformed state-of-the-art single-task and multi-stage models. For image synthesis, our model achieved a PSNR of 25.61 dB, compared to 23.80 dB from a single-task CDM. For anatomical segmentation, it improved the lumen Dice score to 0.89 from 0.87 and the challenging thrombus Dice score to 0.53 from 0.48 (nnU-Net). These segmentation enhancements led to more accurate clinical measurements, reducing the lumen diameter MAE to 4.19 mm from 5.78 mm and the thrombus area error to 33.85% from 41.45% when compared to nnU-Net. Code is available at https://github.com/yuxuanou623/AortaDiff.git.

Figures (2)

• Figure 1: Unified Multitask Framework for Contrast-Free AAA Assessment. This figure illustrates the clinical motivation, CT imaging concepts, and advantages of our proposed multitask diffusion model (AortaDiff) over traditional approaches. (a) The proposed AI-enhanced clinical workflow enables diagnosis from a NCCT scan, eliminating the need for intravenous (IV) contrast agents. (b) A visual comparison highlights the diagnostic challenge: the aortic lumen is obscured in the NCCT but is clearly delineated in the CECT. (c) Our unified multitask model is contrasted with prior methods. Multistage pipelines risk accumulating synthesis errors into the final segmentation. Single-task approaches can produce segmentations that are inconsistent with the synthesized image. In contrast, our end-to-end model jointly generates a high-fidelity synthetic CECT and an accurate, consistent segmentation mask.
• Figure 2: Qualitative Comparison of Synthetic CECT Generation and Lumen and Thrombus Segmentation. This figure illustrates two representative cases comparing generation-only (two-stage pipeline), segmentation-only, and our proposed multi-task diffusion methods. For each case, the top row displays the full generated image, while the bottom row provides a zoomed-in view of the aortic region with the corresponding segmentation result. In both cases, our multi-task methods achieve the most accurate segmentation with clearly defined boundaries and precise localization. Note that for generation-only methods, the segmentation is performed by nnU-Net on the generated synthetic CECT. For segmentation-only methods, no synthetic CECT is produced; thus, the corresponding CT images are shown in gray.